1

Internationally Collaborative Joint

OSSEs Progress At NOAA

Michiko Masutani[1,2,#], Lars Peter Riishojgaard [2,$], Zaizhong Ma[2,$],

Jack S. Woollen[1,+], Dave Emmitt[5], Sid Wood[5], Steve Greco[5],

Tong Zhu[3,@], Yuanfu Xie[4]

[1]NOAA/National Centers for Environmental Prediction (NCEP)

[2]Joint Center for Satellite and Data Assimilation (JCSDA)

[3]NOAA/ NESDIS/STAR,

[4]NOAA/Earth System Research Laboratory (ESRL)

[5]Simpson Weather Associates

# Wyle Information Systems, McLean, VA,

+IM Systems Group)IMSG), MD

$Earth System Science Interdisciplinary Center, Univ. of Maryland, College Park,,

@Cooperative Institute for Research in the Atmosphere (CIRA)/CSU, CO

OSSE:Observing Systems Simulation Experiments

http://www.emc.ncep.noaa.gov/research/JointOSSEs/

Data impact on analysis and forecast will be evaluated.

A Full OSSE can provide detailed quantitative evaluations of

the configuration of observing systems.

A Full OSSE can use an existing operational system and help

the development of an operational system

Advantages

Existing Data assimilation system

and verification method are used for

Full OSSEs. This will help

development of DAS and verification

tools.

OSSE Calibration

A Nature Run (NR, proxy true atmosphere) is produced from a

free forecast run using the highest resolution operational model

which is significantly different from the NWP model used in Data

Assimilation Systems.

Calibrations is performed to provide quantitative data impact

assessment.

. Without calibration quantitative evaluation of data impact is

not possible.

Full OSSEs

There are many types of simulation experiments. Sometimes, we have to call our

OSSE a ‘Full OSSE’ to avoid confusion.

• Full OSSEs are expensive – Sharing one Nature Run and simulated observation saves costs – Sharing diverse resources

• OSSE-based decisions have international stakeholders – Decisions on major space systems have important scientific, technical, financial and political ramifications – Community ownership and oversight of OSSE capability is important for maintaining credibility

• Independent but related data assimilation systems allow us to test the robustness of answers

International Joint OSSE capability

Calibration of OSSEs verifies the simulated data impact by comparing it to real data impact. In order to conduct an OSSE calibration, the data impact of existing instruments has to be compared to their impact in the OSSE.

2

Note: This data must not be used for commercial purposes and re-distribution rights are not given. User lists

lists are maintained by Michiko Masutani and ECMWF

ECMWF Nature run used at NOAA Spectral resolution : T511

13 month long. Starting May1st,2005 Vertical levels: L91, 3 hourly dump

Daily SST and ICE: provided by NCEP Model: Version cy31r1

Based on discussion with JCSDA, NCEP, GMAO, GLA, SIVO, SWA, NESDIS, ESRL,

and ECMWF

Andersson, Erik and Michiko Masutani

2010: Collaboration on Observing

System Simulation Experiments (Joint

OSSE), ECMWF News Letter No. 123,

Spring 2010, 14-16.

Simulated radiance data,

with and without MASK in BUFR format for entire

Nature run period

Type of radiance data and location used for

reanalysis from May 2005-May2006

Simulated using CRTM1.2.2

No observational error added

Conventional data Entire Nature run Period Restricted data removed Cloud track wind is based on real observation location No observational error added

NASA/NCCS

http://portal.nccs.nasa.gov/osse/index.pl

ID and Password required

http://portal.nccs.nasa.gov/josse/index.pl

Ellen Salmon Ellen.M.Salmon@NASA.gov

Bill McHale wmchale@nccs.nasa.gov

NCAR

Currently saved in HPSS

Data ID: ds621.0

Contact:

Chi-Fan Shih chifan@ucar.edu

Steven Worley worley@ucar.edu

Archived in the MARS system at ECMWF

Accessed by external users. Currently available

internally as expver=etwu

3

Simulated observation Control data: Observation type and distribution used by reanalysis for 2005. Observational error is not added to the control data but calibration was performed to demonstrate the impact of observational error in control data. DWL data: GWOS concept DWL simulated by Simpson weather associates.

Case Study to compare impact of DWL with model resolution Atlantic Hurricane in the nature run for the analysis

period of 9/25-10/10

Telescope Modules (4)

Nadir

Star Tracker

• The coherent subsystem provides very accurate (< 1.5m/s) observations when sufficient aerosols (and clouds) exist.

• The direct detection (molecular) subsystem provides observations meeting the threshold requirements above 2km, clouds permitting.

OSSE to evaluate Impact of GWOS DWL GWOS: Global Wind Observing Sounder

4

Related presentations and a poster Thursday, 26 January 2012: 1:45 PM Impact of Different Wind Lidar Configurations on NCEP Forecast Skill Room 340 and 341 (New Orleans Convention Center ) Zaizhong Ma et al. Tuesday, 24 January 2012: 11:30 AM Observing System Simulation Experiments in the Joint Center for Satellite Data Assimilation Room 256 (New Orleans Convention Center ) Lars Peter Riishojgaard et al. Wednesday, 25 January 2012 Joint OSSEs at NOAA, Evaluation of DWL, JPSS, and DWSS Hall E (New Orleans Convention Center ) Michiko Masutani et al

Fig.1Average 500-hPa geopotential height anomaly correlation as a function of forecast range for the Northern (a) and Southern (b) Hemisphere. Tropical wind vector RMS errors (m/s) at 200 hPa (c) and 850 hPa (d) as a function of forecast range. CTRL shown in black, NOUV in red. All observations used were real. Lower plot of each panel shows difference between NOUV and CRTL with, error bars indicating differences that are significant at the 95% confidence level.

Fig. 2: As Figure 1, except that all observations were simulated based on T511NR.

Fig. 3: Average 500-hPa geopotential height anomaly correlation as a function of forecast range in the Northern (a) and Southern (b) Hemisphere. Tropical wind vector RMS errors (m/s) as a function of forecast range at 200 hPa (c) and 850 hPa (d). CTRL in black, NOUV in red, NONW in green, and DWL in blue. Lower plot of each panel shows differences between CTRL and the perturbation experiments (NOUV, NONW or DWL); error bars indicate differences significant at the 95% confidence level. All forecasts were verified against T511NR.

Calibration experiments and DWL OSSE at JCSDA

Calibration and initial evaluation of DWL impact were conducted for the period 1st Jyly-15 August. No observational were added to simulated control data. Calibration experiments showed reasonable agreement in data impact of RAOB wind in real and simulated impact. (Fig.1 and Fig.2) Fig.3 shows GWOS DWL may have more than RAOB wind.

NOUV: NO RAOB wind, NONW: No wind data DWL: CTL + GWOS type DWL

5

Nature run Truth

T126

T170

T382

T254

T126 With DWL

T170 With DWL

T382 With DWL

T254 WithDWL

Minimum Mean Sea level Pressure The verification period:

Sep28-Oct13, 2005 in 72 hour forecast

Evaluated at 00Z only

This display indicates the hurricane track and intensity

6

T170 No obs error in radiance

T170 with large obs error in radiance

Minimum Mean Sea level Pressure The verification period Sep28-Oct13, 2005

72 hour forecast evaluated at 00Z only

T170 with large obs error in radiance

DWL added

T170 no obs error in radiance DWL added

Evaluation of observational error in radiance

Better radiance data help track and intensity forecast. DWL also will improve intensity forecast s even with perfect radiance data.

Imp

rove rad

iance d

ata

Add DWL

Add DWL

Imp

rove rad

iance d

ata

7

Improvement by Adding GWOS DWL

T126

T170

T254

T382

T126 T170

Improvement by Increasing resolution

T170 T254

T254 T382

Improvement by adding GWOS DWL to radiance data with large obs error

Improvement by removing obs error of radiance data. Blue indicate positive impact of random error.

The verification period Sep28-Oct13, 2005 in 72 hour forecast

Evaluated at 00Z only

Impact of resolution vs. GWOS DWL

More verification planned. Add forecast from 12z. Try DWL with other configuration. Produce hurricane track diagnostics.

8

Summary and discussion OSSE with control observation without observational error is useful to provide initial outlook of the data impact in large scale. Adding random error can have a positive impact. DWL improves both intensity and location of a hurricane at all resolution eve with perfect control observation. Adding DWL is more effective than increasing model resolution in Spring Hemisphere. In Northern hemisphere, increasing model resolution will be more effective in large scale forecast. Improvement due to adding DWL is mainly over hurricane. At least T170 resolution is required to utilize DWL data for hurricane forecast . Impact of DWL is larger in T254 than in T170 model forecast but reduced in T382 model forecast with T511 Nature run.

Acknowledgement The nature runs for Joint OSSEs were produced by Dr.

Erik Andersson of ECMWF. We appreciate GMAO to

providing initial satellite data for calibration at ESRL.

GMAO also provided code to add random error to

simulated data.

Future Plans Add various observational errors to control observations and study data sensitivity to the data impact . More OSSEs to study detailed evaluation of configurations of DWL planned by NASA and compared with ESA DWL. Prepare control data for OSSE period with 2011-2012 template. Conduct OSSE to evaluate JPSS and DWSS.

9